The Science of the Total Environment 273 Ž2001.

1᎐10

Effects of electromagnetic fields produced by

radiotelevision broadcasting stations on the immune
system of women

P. Boscolo a,U , M.B. Di Sciascio a , S. D’Ostilio b, A. Del Signore c ,

M. Reale d, P. Conti d, P. Bavazzano e, R. Paganelli a , M. Di Gioacchino a
a
Department of Medicine and Science of Ageing, Uni¨ ersity ‘G. D’Annunzio’, Via dei Vestini, I-66100 Chieti, Italy
b
Agenzia Regionale per la Tutela dell’Ambiente (ARTA), Pescara, Italy
c
Department of Sciences, Uni¨ ersity ‘G. D’Annunzio’, Via dei Vestini, 66100 Chieti, Italy
d
Department of Neurosciences and Oncology, Uni¨ ersity ‘G. D’Annunzio’, Via dei Vestini, 66100 Chieti, Italy
e
Centro di Tossicologia Industriale, Azienda USL 10 r A, Firenze, Italy

Received 9 October 2000; accepted 1 March 2001

Abstract

The object of this study was to investigate the immune system of 19 women with a mean age of 35 years, for at
least 2 years Žmean s 13 years. exposed to electromagnetic fields ŽELMFs. induced by radiotelevision broadcasting
stations in their residential area. In September 1999, the ELMFs Žwith range 500 KHz᎐3 GHz. in the balconies of
the homes of the women were Žmean " S.D.. 4.3" 1.4 Vrm. Forty-seven women of similar age, smoking habits and
atopy composed the control group, with a nearby resident ELMF exposure of - 1.8 Vrm. Blood lead and urinary
trans᎐trans muconic acid Ža metabolite of benzene., markers of exposure to urban traffic, were higher in the control
women. The ELMF exposed group showed a statistically significant reduction of blood NK CD16q-CD56q, cytotoxic
CD3y-CD8q, B and NK activated CD3y᎐HLA-DRq and CD3y-CD25q lymphocytes. ‘In vitro’ production of IL-2
and interferon-␥ ŽINF-␥. by peripheral blood mononuclear cells ŽPBMC. of the ELMF exposed group, incubated
either with or without phytohaemoagglutinin ŽPHA., was significantly lower; the ‘in vitro’ production of IL-2 was
significantly correlated with blood CD16q-CD56q lymphocytes. The stimulation index ŽS.I.. of blastogenesis Žratio
between cell proliferation with and without PHA. of PBMC of ELMF exposed women was lower than that of the
control subjects. The S.I. of blastogenesis of the ELMF exposed group Žbut not blood NK lymphocytes and the ‘in
vitro’ production of IL-2 and INF-␥ by PBMC. was significantly correlated with the ELMF levels. Blood lead and

U
Corresponding author. Tel.: q39-0871-3556704; fax: q39-0871-3556704.
E-mail address: boscolo@unich.it ŽP. Boscolo..

0048-9697r01r$ - see front matter 䊚 2001 Elsevier Science B.V. All rights reserved.
PII: S 0 0 4 8 - 9 6 9 7 Ž 0 1 . 0 0 8 1 5 - 4
2 P. Boscolo et al. r The Science of the Total En¨ ironment 273 (2001) 1᎐10

urinary trans᎐trans muconic acid were barely correlated with immune parameters: the urinary metabolite of benzene
of the control group was only correlated with CD16q-CD56q cells indicating a slight effect of traffic on the immune
system. In conclusion, this study demonstrates that high frequency ELMFs reduce cytotoxic activity in the peripheral
blood of women without a dose᎐response effect. 䊚 2001 Elsevier Science B.V. All rights reserved.

Keywords: Electromagnetic field; Immune system; Lymphocyte subpopulations; Cytokines; Traffic

1. Introduction ‘in vitro’ to low frequency ELMFs showed changes

Several studies have reported that electromag-
netic fields ŽELMFs. modify the human genome
and induce malignancies ŽTheriault, 1996.. An
increased incidence of cancer has been found in
the area surrounding a radio and television trans-
mitter in Great Britain ŽDolk et al., 1997a.. On
the other hand, these results have not been con-
firmed by investigations performed on popula-
tions resident near other power transmitters in
the same country ŽDolk et al., 1997b.. Although
there is concern for the risk of cancer, until now,
a clear relation between ELMF exposure and the
incidence of neoplasms has not been demon-
strated as the studies showed either contradictory
results or the presence of confounding factors
ŽTheriault, 1996; Knave, 2000.. It is clear that
more research on ELMF-exposed populations is
needed to get satisfactory epidemiological study
results or to demonstrate an involvement of
ELMFs in mechanisms which may induce cancer.
Several studies report that the immune system
exerts defence mechanisms against cancer: natu-
ral killer ŽNK. cell activity is reduced in patients
with neoplastic diseases ŽPross and Lotzova, 1993.;
cytotoxic activity against neoplastic cells is an
important prognostic factor in patients with re-
sected lung carcinoma ŽFujisawa and Yamaguchi,
1997.; modern therapy of cancer includes treat-
ment with lymphokines such as interleukin ŽIL. 2
and interferon ŽINF. ␣ and ␥ ŽKamamura et al.,
1998.. For this reason, it may not be excluded
that ELMFs may increase the incidence of cancer
by impairing immune defences.
It was shown that ELMFs modify calcium fluxes
in the membranes of immune cells of humans
acting on the release of tromboxane B 2 and IL-1
ŽConti et al., 1985.. Moreover, peripheral blood
mononuclear cells ŽPBMC. of humans exposed
in w 3 Hxthymidine incorporation following ‘in vitro’
stimulation by mitogens ŽConti et al., 1983, 1986..
Recently, PBMC were exposed for 24, 48 and 72
h to an ELMF with repetition frequency of 50 Hz
and field intensity of 1.5 mT ŽDi Gioacchino et
al., 2000.. The ELMF exposure influenced CD3q,
CD4q, CD8q and CD16q surface marker expres-
sion andror localisation. Moreover, DNA CD4q
expression strongly increased in the exposed cells.
‘In vitro’ cell proliferation of PBMC of subjects
exposed to ELMFs produced by an Italian TV
broadcasting station was different from that of
the control group ŽGiuliani et al., 1996.. More-
over, cytotoxicity tests showed a significant reduc-
tion of NK activity when PBMC of the exposed
group were re-irradiated ‘in vitro’ by radiofre-
quencies Ž639.25 MHz, 12 Vrm average, 50 Hz
amplitude modulated or an ELMF of 50 Hz, 0.67
mT.. We studied immune parameters of men and
women employed in a museum who were exposed
to an ELMF Žrange 0.2᎐3.6 ␮T and 40᎐120 Vrm.
induced by 50 Hz electricity for 20 h a week
ŽBoscolo et al., 2001.. Men working in the museum
showed, in relation to the control subjects, a
statistically significant reduction of both number
and percentage of NK CD16q-CD56q and NK
and B CD3y-CD25q lymphocyte subsets; women
showed a significant reduction in the percentage
of CD3y-CD25q lymphocytes and a slight reduc-
tion of CD16q᎐56q lymphocytes. They also
showed significantly lower levels of INF-␥ in
serum or produced ‘in vitro’ by PBMC, both
spontaneously and stimulated by PHA, while they
showed no significant changes of serum and ‘in
vitro’ produced IL-4, as well as of blastogenesis of
PBMC. These results were similar to the reduced
NK and lymphokine activated activity found in
subjects with poor lifestyle ŽMorimoto et al., 1999,
 P. Boscolo et al. r The Science of the Total En¨ ironment 273 (2001) 1᎐10
2001. or with a psychological behaviour charac-
terised by difficulty in expressing feelings ŽDe-
waraja et al., 1997..
The purpose of this study was to investigate
effects of ELMFs induced by radiotelevison
broadcasting stations on the immune system of
the ELMF exposed population.
2. Subjects and methods
Women were investigated because they re-
mained in their homes on a hill near several
radiotelevision broadcasting stations for a longer
period Ž- 12 h a day. than men. This hill, San
Silvestro, inhabited by approximately 2000 people,
faces the Adriatic Sea 2 km away from Pescara, a
town in Central Italy. During October 1997, the
higher ELMF levels determined in San Silvestro
by the Italian Istituto Superiore Prevenzione Si-
curezza e Lavoro ŽISPESL. and by the regional
environmental protection agency ŽARTA., were
ranging from 11 to 40 Vrm and from 0.5 to 4.0
Wrm2 ŽVignati et al., 1997.; in the same locality
in September 1999, the ELMFs determined by
ARTA ranged from 10 to 25 Vrm. In the period
1997᎐1999, the values of ELMFs in the nearby
towns of Pescara and Chieti Žwhere the control
group was resident., determined by ISPESL and
ARTA, were - 1.8 Vrm. The two towns Žwith a
distance of 15 km. are enclosed between the
Adriatic Sea and the Appennino Mountains. In
this area, high levels of traffic are inside Pescara
and Chieti and in high-speed motorways connect-
ing these towns and those nearby. However, the
toxic compounds produced by traffic and those
emitted in lesser amounts by factories are almost
uniformly distributed in the environment because
of climatic and geographic conditions; the popula-
tion show only slight differences in blood trace
elements, including lead, mainly derived from te-
traethyl and tetramethyl lead added to gasoline
ŽBoscolo et al., 1993..
In September 1999, the values of ELMFs in-
duced by the radiotelevision broadcasting stations
were determined in the balconies and inside the
homes of the women resident in San Silvestro
recruited for this study Žin addition to the values
3
above reported determined in the areas with high
ELMF exposure.. The following instruments
ŽPMM, Milano, Italia . were utilised:
1. probes for electric fields BA-01 Žrange 500
KHz᎐3 GHz; f.s. 120 Vrm.;
2. probes for electric fields BA-05 Žrange 10
MHz᎐1 GHz; f.s. 30 Vrm.;
3. monitor 8051 with optic repeater OR-01;
4. cables with optic fibres to control the instru-
ments at a distance Žto avoid interference of
the operators on ELMF levels.; and
5. wood supports to insulate the probes.
Nineteen women with a mean age of 35 years
Žrange 22᎐49 years. resident in San Silvestro at
least from 1997 Žmean period of residence s 13
years. were investigated. Most of them were
housewives; six Ž31.5%. were smokers Žless than
10 cigarettes a day.; seven Ž36.8%. were atopics:
six had suffered from seasonal asthma andror
rhinitis with serum IgE ) 100 IUrml ŽCAPFeia,
Pharmacia, Uppsala, Sweden. and one of them,
with 92 IUrml of serum IgE, suffered from sea-
sonal rhinitis with positive skin test to inhalant
allergens.
The control group, in which blood lead, serum
IgE and lymphocyte subsets were determined,
was composed of 47 women with a mean age of
35 years Žrange 21᎐49 years.. This group included
another subset of 17 women with a mean age of
34 years Žrange 22᎐44 years., in which urinary
trans᎐trans muconic acid and ‘in vitro’ lym-
phoproliferation and production of cytokines by
PBMC was also measured. The control group of
47 subjects comprised 17 atopics Ž36.1%. and 14
smokers Ž29.8%.; the control group of 17 subjects
was also composed by six atopics Ž35.3%. and five
smokers Ž29.4%.. The diagnosis of atopy was made
considering the levels of serum IgE, history of
allergic disease and skin test to inhalant aller-
gens. Therefore, the ELMF exposed and the con-
trol groups of 47 and 17 women showed similar
age, atopy and smoking habit, while the control
groups, resident in the above reported area of
Pescara and Chieti, showed a higher exposure to
traffic. The exposure to ELMFs produced by elec-
tric appliances in the homes ŽMader and Peralta,
4 P. Boscolo et al. r The Science of the Total En¨ ironment 273 (2001) 1᎐10
1992. of all the women recruited in this study was
not considered.
Clinical assessment included physical examina-
tion and standard routine blood analyses. Preg-
nant women, those taking drugs or suffering from
diseases were not recruited for the investigation.
Atopic women in the active phase of allergic
disease treated with systemic drugs Žsteroids
andror antihistamines . were also excluded.
Informed consent was obtained from the re-
cruited subjects according to a procedure ap-
proved by the ‘Ethic committee’ of the University
‘G. D’Annunzio’ of Chieti. Blood and urine sam-
ples were collected in plastic cryovials ŽNalgene,
International PBI, Milano, Italy. at 08.00 h, with
a standard procedure ŽSabbioni et al., 1992. in
order to determine blood lead and urinary
trans᎐trans muconic acid Ža metabolite of ben-
zene., markers of exposure to traffic ŽBoscolo et
al., 2000.. Blood lead was determined by an atomic
absorption spectrophotometer Perkin-Elmer 4100
ZL ŽSabbioni et al., 1992.. A further analytical
quality control on 10% of samples was carried out
in two different laboratories. Urinary trans᎐trans
muconic acid was analysed by HPLC ŽDucos et
al., 1990, 1992..
Fluorescein isothiocyanate ŽFITC. and phy-
coerythrin ŽPE.-conjugated antibodies ŽBecton-
Dickinson, San Jose, CA, USA. were used to
determine lymphocyte subsets. The antibodies
were CD4-CD45RO wto evaluate helper CD4q-
CD45ROq ‘memory’ and CD4q-CD45ROy ‘naive
or virgin’ lymphocytes ŽMale et al., 1996.x, CD3-
CD8, CD16-CD56 ŽNK cells., CD19 ŽB lympho-
cytes., CD3-HLADR and CD3-CD25 Žboth acti-
vated NK and B lymphocytes.. Two-colour flow-
cytometry analysis was performed by FACscan
ŽBecton-Dickinson, San Jose, CA, USA. ŽFleiscer
et al., 1988.. Serum IgE was measured by ELISA
ŽBrostoff et al., 1991..
For determining the production of cytokines,
PBMC were incubated for 24 h at 37⬚C in 5%
CO 2-humidified atmosphere in polypropylene
tubes ŽFalcon, Italy. with or without 20 ␮grml of
phytohaemoagglutinin ŽPHA. ŽDifco.. At the end
of the incubation period, cell-free supernatants
were harvested and stored at y20⬚C until assay
of IL-2, IL-4, IL-5 and INF-␥ ŽBenfer-Scheller,
Key-Stone Laboratories, USA. with or without
PHA by ELISA ŽFridas et al., 1996..
The blastogenesis Žproliferation. of PBMC was
also determined ‘in vitro’ according to Conti et al.
Ž1983, 1986.. Blastogenesis was determined as the
stimulation index ŽS.I.., which is the rate between
w 3 Hxthymidine incorporation by PBMC in the
presence of PHA and without PHA, in the last 6
h of incubation after a 48-h incubation period.
Statistical analysis of the data was performed
with Statistica, Release 4.5. Kolmogorov᎐Smirnov
tests showed that most of the results concerning
immune parameters did not conform to the nor-
mal distribution. Therefore, non-parametric
methods were used for all simple descriptive
statistics.
3. Results
ELMFs with range 500 KHz᎐3 GHz, de-
termined during the mornings of September 1999
in the balconies of the homes of 17 out of the 19
investigated women resident in San Silvestro hill
were Žmean " S.D.. 4.3 " 1.4 Vrm. ELMFs,
within the same range of 500 KHz᎐3 GHz, de-
termined inside the home, were approximately
30᎐50% lower than outside. ELMF values showed
slight modifications during the 24-h. The levels of
ELMFs of the homes of the recruited women
were lower than those Žranging from 10 to 24
Vrm reported in Section 2. determined in an-
other area of San Silvestro. Only one ELMF
exposed woman showed values of ELMF expo-
sure Ž1.2 Vrm outside her home. similar to that
of the control group Ž- 1.8 Vrm..
The values of blood lead and urinary trans᎐trans
muconic acid, biomarkers, of exposure to traffic
ŽBoscolo et al., 1999a, 2000. of the women of San
Silvestro were lower than those of the control
group resident in the nearby area of the towns of
Pescara and Chieti ŽTable 1..
Serum IgE of the ELMF exposed women did
not show statistically significant differences from
those of the control subjects ŽTable 2..
ELMF exposed women showed a significant
reduction of T ‘virgin’ CD4q-CD45ROy lympho-
cytes and a higher ratio between ‘memory’ CD4q-
 P. Boscolo et al. r The Science of the Total En¨ ironment 273 (2001) 1᎐10 5

Table 1
Blood metals and urine trans᎐trans muconic acid of women exposed to electromagnetic fields a

Control Exposed
No. median 25th᎐75th No. Median 25th᎐75th
percentiles percentiles

Blood lead Ž␮grl. 47 5.9 4.9᎐7.0 19 4.4 3.7᎐4.7U

Urinary trans᎐trans-muconic 17 42.0 19.1᎐107.5 19 15.0 9.5᎐32.5U
acid Ž␮grl.
Urinary trans᎐trans-muconic 17 33.5 13.2᎐92.5 19 11.0 6.5᎐28.0UU
acid Ž␮grg creatinine .
a
Mann᎐Whitney U-test. Statistical significant difference: U P- 0.05; UU P - 0.01.

CD45ROq and ‘virgin’ CD4q-CD45ROy lym-
phocytes ŽTable 2.. ELMF exposed women showed
a reduction in both the number and percentage of
total lymphocytes of blood NK CD16q-CD56q
lymphocytes, CD3y-CD8q cells wCD8q lympho-
cytes without the expression of the antigen T
CD3q and NK CD16q cells ŽMale et al., 1996.x
and of CD3y-CD25q B and NK activated lym-
phocytes ŽTable 2 and Fig. 1.. CD3y-HLA-DRq
B and NK activated lymphocytes of the ELMF
exposed group were also significantly lower than
those of the control group when determined as
the percentage of total lymphocytes ŽTable 2 and
Fig. 1..
In ‘vitro’ production of di IL-2 and INF-␥ by
PBMC of the ELMF exposed women was signifi-
cantly lower than that of the control group either
without or with PHA in the incubation liquid
ŽTable 3.. On the other hand, there was no dif-
ference in the production of IL-4 and IL-5 by
PBMC of the ELMF exposed and control women
ŽTable 3..

Table 2
Serum IgE and blood lymphocyte subpopulations of women exposed to electromagnetic fields a

Cellsr␮l Control Ž n 47. Exposed Ž n 19.

median 25th᎐75th median 25th᎐75th
percentiles percentiles

Serum IgE ŽIUrml. 25 10᎐79 41 14᎐121

Lymphocytes 1999 1780᎐2605 2050 1700᎐2545
CD3q 1421 1270᎐1811 1531 1214᎐1854
CD4q 884 711᎐1213 955 684᎐1152
CD4q-CD45ROy 323 260᎐477 264 211᎐347U
CD4q-CD45ROq 562 448᎐711 673 496᎐779
CD4q-CD45ROqrCD4q-CD45ROy 1.50 1.01᎐2.02 2.36 1.99᎐3.18U
CD3q-CD8q 520 442᎐726 562 467᎐674
CD3y-CD8q 133 98᎐194 93 56᎐123UU
CD16q-CD56q 395 306᎐489 239 176᎐293UU
CD19q 210 138᎐291 208 153᎐251
HLA-DRq 482 392᎐601 432 361᎐561
CD3y᎐HLA-DRy 330 263᎐431 295 250᎐355
CD3q᎐HLA-DRq 124 92᎐176 131 89᎐220
CD25q 356 273᎐496 372 263᎐444
CD3y-CD25y 95 77᎐126 68 31᎐92UU
CD3q-CD25q 248 195᎐351 287 239᎐386
a
Mann᎐Whitney U-test. Statistical significant difference: U P- 0.05; UU P - 0.01.
6 P. Boscolo et al. r The Science of the Total En¨ ironment 273 (2001) 1᎐10

Fig. 1. Blood CD16q-CD56q, CD3y-HLA-DRq, CD3y-CD25q and CD3y-CD8y lymphocyte subsets Žexpressed as percentage of
total lymphocytes. of women exposed to electromagnetic fields. Values are expressed as mean " S.E. Mann᎐Whitney U-test
Žcontrol women: n s 47; ELMF exposed women: n s 19.. Statistical significant difference: 䢇 P- 0.05; 䢇 䢇 P - 0.01.

The S.I. of blastogenesis Žproliferation. of
PBMC, determined using PHA, of the ELMF
exposed group was significantly lower than that of
the control group ŽTable 4.. Notwithstanding this
result, there was a significant correlation between
the values of ELMFs determined from the bal-
conies of 17 homes of the women resident in San
Silvestro and their S.I. of blastogenesis. The curve
which better interpolates data regarding the rela-
tion between Vrm and S.I. is illustrated in Fig. 2:
it is a logarithmic curve having the following
equation: y s 16.944 x 0.553 with R 2 s 0.5519. On
the other hand, there was no significant correla-
tion between the values of ELMFs determined in
the homes of the women resident in San Silvestro
and their lymphocyte subsets or production of
cytokines by their PBMC.
Blood lymphocytes CD16q-CD56q Žreported as
percentage of total lymphocytes. of all the ex-
amined women showed a positive linear correla-
tion with the values of IL-2 produced ‘in vitro’ by
PBMC either incubated in absence of PHA Ž n s
36, R s 0.557, P- 0.001. or with PHA Ž n s 36,
R s 0.416, P- 0.02.. Moreover, INF-␥ produced
‘in vitro’ by PBMC with PHA was significantly
correlated with the IL-2 produced ‘in vitro’ by

Table 3
‘In vitro’ production of cytokines by peripheral blood mononuclear cells ŽPBMC. incubated with or without phytohaemoagglutinin
ŽPHA. of women exposed to electromagnetic fields a

Cytokines Žpgrml. Control Ž n 17. Exposed Ž n 19.

median 25th᎐75th median 25th᎐75th
percentiles percentiles

IL-2 without PHA 17.70 5.50᎐38.70 1.57 1.29᎐2.50UU

IL-2 with PHA 55.30 17.01᎐163.20 6.70 3.23᎐32.80U
IL-4 without PHA 1.70 1.29᎐3.01 2.11 1.5᎐2.59
IL-4 with PHA 3.30 1.60᎐5.80 3.31 2.35᎐5.40
IL-5 without PHA 1.28 0.85᎐1.98 1.90 1.10᎐3.27
IL-5 with PHA 12.50 6.95᎐18.10 11.40 6.20᎐37.25
INF-␥ without PHA 0.81 0.49᎐1.11 n.d.
INF-␥ with PHA 27.50 8.30᎐51.50 7.60 1.03᎐15.65U
a
Mann᎐Whitney U-test. Statistical significant difference: U P- 0.002; UU P- 0.001; n.d.s non-determinable.
 P. Boscolo et al. r The Science of the Total En¨ ironment 273 (2001) 1᎐10
Table 4
Stimulation index ŽS.I.. a of blastogenesis Žproliferation. of peripheral blood mononuclear cells ŽPBMC. of women exposed to
electromagnetic fields b
Control Ž n 17.
median 25th᎐75th
percentiles
Simulation index 54.5 37.6᎐71.4
ŽS.I..
a
S.I. is the rate between w 3 Hxthymidine incorporation by PBMC in presence of phytohaemoagglutinin ŽPHA. and without PHA.
b
Mann᎐Whitney U-test. Statistical significant difference: U P- 0.05.
PBMC both incubated in absence of PHA Ž n s 36,
R s 0.674, P- 0.0001. and with PHA Ž n s 36,
R s 0.410, P- 0.02..
Blood lead and urinary trans᎐trans muconic
acid did not show a significant correlation with
immune parameters with the exception of the
correlation between urinary trans᎐trans muconic
acid Žreported as ␮grg creatinine . and CD16q-
CD56q lymphocytes Ž n s 17, R s 0.627, P- 0.01.
and CD4q-CD45ROy ‘virgin’ lymphocytes Ž n s
17, R s 0.625, P- 0.01. of the control group. For
this reason, a correlation analysis between ELMF
values of exposure and immune parameters Žin-
cluding CD16q-CD56q, CD3y-HLA-DRq, CD3y-
CD25q, CD3y-CD8q cells, S.I. of blastogenesis of
PBMC and production of INF-␥ and IL-2 by
Fig. 2. Linear correlation between the values of electromag-
netic fields produced by radiotelevision broadcasting stations
on the balconies of the homes of women and their stimulation
index ŽS.I.. of blastogenesis Žproliferation. of peripheral blood
mononuclear cells ŽPBMC.. Pearson’s correlation coefficient:
n s 17, R s 0.608; P- 0.01.
7
Exposed Ž n 19.
median 25th᎐75th
percentiles
36.8 32.2᎐43.2U
PBMC. was performed with adjustment for uri-
nary trans᎐trans muconic acid: all these statistical
analyses were not significant.
4. Discussion
The women exposed to ELMFs produced by
radiotelevision broadcasting stations showed a re-
duction of NK CD16q-CD56q and cytotoxic
CD3y-CD8q cells and of NK and B CD3y-HLA-
DRq and CD3y-CD25q activated lymphocytes in
the peripheral blood along with a significant re-
duction of IL-2 and INF-␥ produced ‘in vitro’ by
PBMC. None of these immune parameters
showed a dose᎐response effect with the ELMF
values determined outside the homes of the in-
vestigated women.
Blood NK CD16q-CD56q lymphocytes of all
the examined subjects were significantly corre-
lated with the values of IL-2 and INF-␥ produced
‘in vitro’ by PBMC. It is known that IL-2 and
INF-␥, produced by activated T lymphocytes, ex-
ert a potent effect on macrophages and a syner-
gistic action with that of other cytokines we.g.
TNF-␣ ŽMale et al., 1996.x. It is also known that
INF-␥ produced by both T and NK lymphocytes,
may activate NK lymphocytes with an ‘autocrine
loop’ ŽMale et al., 1996.. For this reason, lower
levels of blood NK cells and of the ‘in vitro’
production of IL-2 and INF-␥ by PBMC of the
ELMF exposed women are an expression of low
cytotoxic activity in the peripheral blood.
The results of this study are in agreement with
those of Giuliani et al. Ž1996., who showed al-
tered cytotoxicity tests in subjects exposed to
8 P. Boscolo et al. r The Science of the Total En¨ ironment 273 (2001) 1᎐10
ELMFs produced by a TV broadcasting station.
This investigation also confirms the data of a
previous study of our group on men and women
employed in a museum exposed to an ELMF
induced by 50 Hz electricity who showed low
blood NK CD16q-CD56q andror B and NK
CD3y-CD25q lymphocytes and reduced produc-
tion of INF-␥ by PBMC ŽBoscolo et al., 2001.
The results of this investigation are similar to
those of Morimoto et al. Ž1999, 2001. and De-
waraja et al. Ž1997. who found low blood NK
activity or a reduction of blood NK lymphocytes
in subjects with poor lifestyle and with a psycho-
logical behaviour characterised by difficulty in
expressing feelings. Therefore, we do not exclude
that the effects of the ELMFs on the immune
system may be, in part, mediated by those on
nervous mechanisms which are closely connected
with the immune ones ŽMale et al., 1996; Jankovic,
1992; Szentivayi, 1997.. With regard to this, the
blood levels of both total and cytotoxic lympho-
cytes decreased following exposure to magnetic
resonance imaging of the brain of volunteers,
suggesting that the nervous system may regulate
the trafficking of lymphocytes in peripheral blood
ŽReichard et al., 1996.. It was also shown that the
circadian biorhythm regulates not only the re-
lease and production of neurohormones, but also
blood levels of lymphocyte subsets ŽBoscolo et al.,
1999b..
Although blood cytotoxic activity may be modi-
fied by lifestyle ŽMorimoto et al., 1999, 2001.,
psychological behaviour ŽDewaraja et al., 1997.
and possibly occupational stress ŽBoscolo et al.,
2001., the monitoring of blood cytotoxic activity
and NK lymphocytes may be considered useful as
a biomarker of exposure to ELMFs. A more
interesting biomarker seems to be the CD16q-
CD56q NK lymphocyte subpopulation, which is
significantly correlated with the ‘in vitro’ produc-
tion of IL-2 by PBMC.
Blood lead and urinary trans᎐trans-muconic
acid, markers of exposure to traffic ŽBoscolo et
al., 2000. of the control subjects, were higher than
those of the ELMF exposed. These markers of
traffic exposure were not correlated with parame-
ters of cytotoxic activity, with the exception of the
correlation between urinary trans᎐trans-muconic
acid and CD16q-CD56q lymphocytes of the con-
trol group. We could thus suggest that the dif-
ference in the values of CD16q-CD56q NK lym-
phocytes, between ELMF exposed and control
women may rather depend on a higher effect of
the toxic compounds produced by traffic on the
control subjects. However, adjustment for urinary
trans᎐trans-muconic acid did not show significant
correlation between ELMF values and immune
parameters Žincluding CD16q-CD56q lympho-
cytes.. For this reason, an effect of the exposure
to traffic on the values of NK cells has to be
considered as unimportant.
Urinary trans᎐trans-muconic acid was also
positively correlated with CD4q-CD45ROy ‘virgin
¨ lymphocytes of the control group. More-
or naıve’
over, the control group showed higher blood lev-
els of CD4q-CD45ROy lymphocytes, as well as of
lead, than the ELMF exposed group. Sata et al.
Ž1998. found a correlation between the percent-
¨ lymphocytes and blood lead of male
age of ‘naıve’
workers with mean blood lead of 19 ugrdl. These
data suggest that the values of blood CD4q-
CD45ROy and their ratio with CD4q-CD45ROq
lymphocytes may depend on several factors, in-
cluding lead and benzene produced by traffic.
ELMF exposed women showed a reduced sti-
mulation of PHA on the proliferation of PBMC
in relation to the control subjects with a higher
exposure to traffic. We cannot exclude an en-
hancement of the proliferation of PBMC induced
by the toxic compounds of traffic, as in the previ-
ous case. On the other hand, the S.I. of blastoge-
nesis of the ELMF exposed women Žalthough
lower than that of the control subjects. was posi-
tively correlated with the ELMF levels de-
termined outside their homes. Therefore, an ef-
fect of ELMFs in stimulating PBMC proliferation
may also not be excluded.
A part of the immune parameters investigated
in this study, plasma zinc and copper markers of
metabolic activities ŽPrasad, 1988; Harbige, 1996.,
blood lead and urinary trans᎐trans-muconic acid
were determined in women exposed to low and
high frequency ELMFs and in control women
considering the presence of atopy ŽDel Signore et
al., 2000.. Linear discriminant analysis applied to
the above mentioned parameters showed no sig-
 P. Boscolo et al. r The Science of the Total En¨ ironment 273 (2001) 1᎐10
nificant difference between non-atopic groups ex-
posed or not exposed to ELMFs, whereas there
was a marked differentiation between atopic
women non-exposed and atopic women ELMF
exposed. This suggested that ELMFs have a
greater influence on atopic women than on non-
atopic ones.
We may conclude that exposure to ELMFs
induces a modification of immune parameters in
humans. ELMF exposure inhibits a Th1-like cyto-
toxic immune-response without a dose᎐response
effect, while the enhancement of a Th2-like im-
mune response is not demonstrated. Follow-up
studies may be useful to explain the biological
importance of the reported results.
Acknowledgements
This investigation was supported by Italian
MURST and ISPESL.
References
Boscolo P, Sabbioni E, Di Giacomo F, Sforza RG, Giaccio M.
Preliminary study on trace elements reference values in
blood and urine from inhabitants of Abruzzo ŽCentral
Italy.. In: Anke M, Mills CF, editors. Trace elements in
men and animals. TEMA 8. Proceedings of TEMA 8, May
1993, Dresden, Verlag Media Touristik, Leipzig, 1993:
422᎐426.
Boscolo P, Di Gioacchino M, Sabbioni E, Benvenuti F, Conti
P, Reale M, Bavazzano P, Guiliano G. Expression of lym-
phocyte subpopulations, cytokine serum levels and blood
and urine trace elements in asymptomatic atopic men ex-
posed to urban environment. Int Arch Occup Environ
Health 1999a;72:26᎐32.
Boscolo P, Andreassi M, Sabbioni E, Reale M, Conti P,
Amerio P, Di Gioacchino M. Systemic effects of ingested
nickel on the immune system of nickel sensitised women.
Life Sci 1999b;64:1485᎐1491.
Boscolo P, Di Gioacchino M, Sabbioni E, Reale M, Volpe
AR, Di Sciascio MB, Conti P, Guiliano G. Lymphocyte
subpopulations, cytokines and trace elements in asympto-
matic atopic women exposed to an urban environment. Life
Sci 2000;67:1119᎐1126.
Boscolo P, Bergamaschi A, Di Sciascio MB, Benvenuti F,
Reale M, Di Stefano F, Conti P, Di Gioacchino M. Effects
of low frequency electromagnetic fields on expression of
lymphocyte subsets and production of cytokines of men and
women employed in a museum. Sci Total Environ 2001 Žin
press..
9
Brostoff J, Scadding GK, Male D, Roitt IM, editors. Clinical
immunology. London᎐New York: Gower Medical Publish-
ing, 1991.
Conti P, Gigante GE, Cifone MG, Alesse E, Ianni GF, Reale
M, Angeletti PU. Reduced mitogenic stimulation on hu-
man lymphocytes by extremely low frequency electromag-
netic fields. FEBS Lett 1983;162:156᎐160.
Conti P, Gigante GE, Alesse E, Cifone MG, Fieschi C, Reale
M, Angeletti PU. A role for Ca2q in the effect of very low
frequency electromagnetic field on the blastogenesis of
human lymphocytes. FEBS Lett 1985;181:28᎐32.
Conti P, Gigante GE, Cifone MG, Alesse E, Fieschi C,
Bologna M, Angeletti PU. Mitogen dose-dependent effect
of weak pulsed electromagnetic field on lymphocyte blasto-
genesis. FEBS Lett 1986;199:130᎐134.
Del Signore A, Boscolo P, Kouri S, Di Martino G, Giuliano G.
Combined effects of traffic and electromagnetic fields on
the immune system of fertile atopic women. Ind Health
2000;38:294᎐300.
Dewaraja R, Tanigawa T, Araki S, Nakata A, Kawamura N,
Ago Y, Sasaki Y. Decreased cytotoxic lymphocyte counts in
alexithymia. Psychother Psychosom 1997;66:83᎐86.
Di Gioacchino M, Conti P, Boscolo P. Immune effects of
electromagnetic fields: in vitro study. III International Sym-
posium on Occupational and Environmental Allergy and
Immune Diseases ’99. Singapore, 27 August 2000. Int J
Immunopathol Pharmacol Žin press..
Dolk H, Shaddick G, Walls P, Grundy C, Thakrar B, Klein-
schmidt I, Elliott P. Cancer incidence near radio and
television transmitters in Great Britain ᎏ the Sutton Cold-
field transmitters. Am J Epidemiol 1997a;145:1᎐9.
Dolk H, Elliott P, Shaddick G, Walls P, Thakrar B. Cancer
incidence near radio and television transmitters in Great
Britain ᎏ all high power transmitters. Am J Epidemiol
1997b;145:1017.
Ducos P, Gaudin R, Robert A, Francin JM, Maire C. Im-
provement in HPLC analysis of urinary trans᎐trans-muconic
acid, a promising substitute for phenol in the assessment of
benzene exposure. Int Arch Occup Environ Health
1990;62:529᎐534.
Ducos P, Gaudin R, Bel J, Maire C, Francin JM, Robert A,
Wild P. Trans᎐trans-muconic acid, a reliable biological indi-
cator for the detection of individual benzene exposure
down to the ppm level. Int Arch Occup Environ Health
1992;64:309᎐313.
Fleiscer TA, Agengruber C, Marti GE. Immunophenoyping of
normal lymphocytes. Pathol Immunopathol Res 1988;
7:305᎐316.
Fridas S, Karagouni E, Dotsika E, Reale M, Barbacane RC,
Vlem-Mas I, Anogiakis GA, Tracatellis A, Conti P. Gener-
ation of TNF␣ , IFN␥, IL-6, IL-4 and IL-10 in mouse serum
from trichinellosis: effect of the anti-inflammatory com-
pound 4-deoxypiridoxine Ž4-DPD.. Immunol Lett 1996;
49:17᎐184.
Fujisawa T, Yamaguchi Y. Autologous tumor killing activity as
a prognostic factor in primary resected nonsmall cell carci-
noma of the lung. Cancer 1997;79:474᎐481.
10 P. Boscolo et al. r The Science of the Total En¨ ironment 273 (2001) 1᎐10
Giuliani L, Vignati M, Cifone MG, Alesse E. Similarity of
effects induced by ELF amplitude modulated RF and ELF
magnetic fields on PHB in vitro. ICOH’96, Int. Cong.
Occup. Health, Stoccolma, 15᎐20 September, 1996:309.
Harbige LS. Nutrition and immunity with emphasis on infec-
tion and autoimmune disease. Nutr Health 1996;10:
285᎐312.
Jankovic BD. The neuro-immune network. Some recent devel-
opment. Rec Progr Med 1992;83:93᎐99.
Kamamura Y, Takahashi K, Homaki K, Monden Y. Effects of
interferon ␣ and ␥ on development of LAK activity from
mononuclear cells in breast cancer patients. J Med Invest
1998;45:71᎐75.
Knave B. Electromagnetic fields and health outcomes. 26th
Int. Cong. Occup. Health, Singapore, 27th August, 1st
September 2000, key-note addresses n 9, 2000:7.7᎐8.4.
Mader DL, Peralta SB. Residential exposure to 6⬚ Hz mag-
netic fields from appliances. Bioelectromagnetics 1992;
11:283᎐296.
Male D, Cooke A, Owen M, Trowsdale J, Champion B.
Advanced immunology, third edition. London: Mosby, 1996.
Morimoto K, Takeshita T, Sakurai C. Lifestyle and immuno-
logical potential. II International Symposium on Occupa-
tional and Environmental Allergy and Immune Diseases
’99. Chieti, 27᎐30 aprile 1999. Int J Immunopathol Phar-
macol 1999;12Ž2s.:32.
Morimoto K, Takeshita T, Inoue-Sakurai C, Maruyama S.
Lifestyle and mental health status are associated with natu-
ral killer cell and lymphokine activated killer cell activities.
Sci Total Environ 2001 Žin press..
Prasad AS. Zinc in growth and development and spectrum of
human zinc deficiency. J Am Coll Nutr 1988;7:377᎐384.
Pross HF, Lotzova E. Role of natural killer cells in cancer.
Nat Immun 1993;12:279᎐292.
Reichard SM, Allison JD, Figueroa RE, Dickinson MM, Reese
AC. Leukocyte trafficking in response to magnetic reso-
nance imaging. Experientia 1996;52:51᎐54.
Sabbioni E, Minoia C, Pietra R, Fortaner S, Gallorini M,
Saltelli A. Trace element reference values in tissues from
inhabitants of the European Community. II. Examples of
strategy adopted and trace element analysis of blood, lymph
nodes and cerebrospinal fluid of Italian subjects. Sci Total
Environ 1992;120:39᎐62.
Sata F, Araki S, Tanigava T, Morita Y, Sakurai S, Nakata A,
Katsuno N. Changes in T cell subpopulations in lead work-
ers. Environ Res 1998;76:61᎐64.
Szentivayi A. The immune neuroendocrine circuity in health
and atopic disease. Progress in Allergy and Clinical Im-
munology, 4. Seattle᎐Toronto᎐Bern: Cancum, Hogrefe and
Huber Publ, 1997:343᎐346.
Theriault G. Electromagnetic fields and cancer: a critical
review of occupational studies. ICOH’96, 25th Int. Cong.
Occup. Health, Stockholm 15᎐20 September 1996, key-note
addresses, 1996:25᎐35.
Vignati M, Giuliani L, Graziani R. Relazione Tecnica. Espo-
` S. Sivestro di Pescara
sizione della popolazione in localita.
ai campi elettromagnetici generati da antenne radiotelevi-
sive. ISPESL, prot. 5234, 1997.